First SDO AIA Observations of a Global Coronal EUV "Wave": Multiple Components and "Ripples"

5 MB·English

by  Wei Liu

The Astrophysical JournalLetters, 723: L53-L59,2010November1
PreprinttypesetusingLATEXstyleemulateapjv.5/2/11
FIRST SDO AIA OBSERVATIONS OF A GLOBAL CORONAL EUV “WAVE”: MULTIPLE COMPONENTS
AND “RIPPLES”
Wei Liu12, Nariaki V. Nitta1, Carolus J. Schrijver1, Alan M. Title1, and Theodore D. Tarbell1
The Astrophysical Journal Letters,723: L53-L59, 2010 November 1
ABSTRACT
We present the first SDO AIA observations of a global coronal EUV disturbance (so-called “EIT
wave”) revealed in unprecedented detail. The disturbance observed on 2010 April 8 exhibits two
2
components: one diffuse pulse superimposed on which are multiple sharp fronts that have slow and
1
fast components. The disturbance originatesin frontof erupting coronalloops andsome sharpfronts
0
2 undergo accelerations, both effects implying that the disturbance is driven by a CME. The diffuse
pulse,propagatingatauniformvelocityof204–238kms−1 withverylittleangulardependencewithin
n its extent in the south, maintains its coherence and stable profile for ∼30 minutes. Its arrival at
a
increasing distances coincides with the onsets of loop expansions and the slow sharp front. The fast
J
sharpfrontovertakesthe slowfront, producing multiple “ripples”andsteepening the localpulse, and
4
both fronts propagate independently afterwards. This behavior resembles the nature of real waves.

Unexpectedly, the amplitude and FWHM of the diffuse pulse decrease linearly with distance. A
]
R hybrid model, combining both wave and non-wave components, can explain many, but not all, of the
observations. Discoveriesofthetwo-componentfrontsandmultipleripplesweremadepossibleforthe
S
first time thanks to AIA’s high cadences (≤20 s) and high signal-to-noise ratio.
.
h Subject headings: Sun: activity—Sun: corona—Sun: coronal mass ejections—Sun: flares—Sun: UV
p radiation
-
o
tr 1. INTRODUCTION (1′.′4, with 0′.′6 pixels). These capabilities allow us to
s study the kinematics and thermal structure of EUV dis-
a Expanding annular, global-scale, coronal EUV en-
turbances in unprecedented detail, as reported in this
[ hancements were discovered with SOHO EIT and
have been often called “EIT waves” (Moses et al. Letter.
1 1997; Thompson et al. 1998). They propagate across 2. OBSERVATIONSANDDATAANALYSIS
5v the solar disk at typical velocities of 200–400kms−1 Theeventofinterest(SolarObjectLocator: SOL2010-
1 (Thompson & Myers 2009), slower than Moreton(1960) 04-08T02:30:00L177C061) occurred in NOAA active re-
8 waves (∼1000 kms−1). Their nature is still un- gion (AR) 11060 (N25E16) from 02:30 to 04:30 UT on
0 der intense debate. Competing models include: fast- 2010April 8. It involveda GOES B3.8 two-ribbonflare,
. mode MHD waves (Wills-Davey & Thompson 1999; CME,andglobalEUVdisturbance,asshowninFigure1.
1
Wang 2000; Wu et al. 2001; Warmuth et al. 2001; No Moreton wave was found in available Hα data (from
0
Ofman & Thompson 2002; Schmidt & Ofman 2010), the Huairou Solar Observing Station, China).
2
slow-mode waves or solitons (Wills-Davey et al. 2007; AIA images (20 s cadence) were first coaligned across
1
Wang et al. 2009), and non-waves related to a cur- wavelengths and differentially rotated to a reference
:
v rent shell or successive restructuring of field lines at time (03:33:47 UT). Figure 2 shows selected images.
i the CME front (Delann´ee 2000; Chen et al. 2002, 2005; To minimize human subjectivity, we employed a semi-
X
Attrill et al. 2007; Attrill 2010). Details of observations automatic approach (cf., Wills-Davey 2006), following
r and models can be found in recent reviews (Warmuth Podladchikova& Berghmans (2005). We identified the
a
2007;Vrˇsnak & Cliver2008;Wills-Davey & Attrill2009; eruption center (flare kernel centroid at x = −238′′,
Gallagher & Long 2010). y =487′′)asthenew“northpole”,anddrew15◦widehe-
Major drawbacks that hindered our understanding of liographic “longitude” sectors (A0–A11, Figures 2(a)–
such propagating EUV disturbances were EIT’s single (b)). For each sector, we obtained the image profile as a
view point and low cadence (≥12 minutes), which were function of distance measured from the eruption center
partiallyalleviatedbySTEREOEUVI(Long et al.2008; along“longitudinal”greatcircles(thuscorrectingforthe
Veronig et al.2008;Ma et al.2009;Kienreich et al.2009; atmosphere’ssphericity)byaveragingpixelsinthe “lati-
Patsourakos& Vourlidas2009). TheAtmosphericImag- tudinal”direction. Composingsuch1Dprofilesovertime
ingAssembly (AIA) onthe SolarDynamicsObservatory givesa2Dspace-timeplot,fromwhichbaseandrunning
(SDO)has10EUVandUVwavelengths,coveringawide ratiosanddifferences(givingdifferentcolorscalingsuited
range of temperatures, at high cadence (up to 10–20 s, for different features) can be readily obtained, as shown
∼10× faster than STEREO’s 75–150 s) and resolution in Figures 3 and 5.
TheglobaldisturbanceisvisibleinallsevenAIAEUV
1Lockheed Martin Solar and Astrophysics Laboratory, De- channels and for the first time observed at 131, 94,
partmentADBS,Building252,3251HanoverStreet, PaloAlto, 335, and 211 ˚A. It is similar in channels >1 MK, be-
CA94304
2W. W. Hansen Experimental Physics Laboratory, Stanford ing the strongest at 193 and 211 ˚A. It appears different
University,Stanford,CA94305 in the cooler 171 ˚A channel and the weakest at 304 and
2 Liu et al.
Fig.1.—RunningdifferenceimagesofcompositeSTEREO-ACOR1andEUVI195˚A(top,seeonlineMovie1)andAIA193˚A(bottom),
where the ecliptic and solar norths are up, respectively. Heliographic grids are spaced by 10◦. Arrows point to the leading front of the
diffusepulseseeninwhitelightandEUV.ThecyanarrowsarealsoapproximatelytheprojectionofAIA’sline-of-sight(LOS;intheecliptic
plane)seenbySTEREO-A.
131˚A.Herewefocusontherepresentative193and171˚A 4. GLOBALEUVDISTURBANCE
channels (hereafter 193 and 171). Incidentally, we no- 4.1. Overview
ticed very fast (500–1200kms−1), narrow-angle features
The global EUV disturbance, seen by AIA on the
launchedsouthwardfromtheeruptingAR,repeatedlyat
∼100 s intervals throughout a 1.5 hrs duration, even af- solar disk, exhibits two components: one broad, dif-
fuse pulsesuperimposed onwhichare multiple narrow,
tertheglobaldisturbancehadleft. Theirnatureisunder
sharp fronts (see Movies 2c–d). The diffuse pulse ap-
investigation and will be presented in the future.
pearsasanannular-shapedemissionenhancementahead
3. ERUPTINGLOOPSANDCME of the erupting loops. It expands in all directions ex-
There is initially a series of coronal loops, best seen cept to the north, being the strongest in the southwest
at 171, fanning out along an arcade in the northwest- (Figure 1, bottom; Figure 2(d)). The sharp fronts also
southeast direction. At the flare onset of 02:32 UT, havetwocomponents: aslow frontthatappearsearlier
these loops, as shown in Figure 2(c), start to expand andoccupiesfromthe westtosouthwestofthe AR(Fig-
southwardto the quietSun, likelyavoidingthe probably ures 2(d)–(g)), and a fast front that comes later near
stronger magnetic field of large scale loops in the north thepeaklocationofthediffusepulseanddominatesfrom
(see Movies 2a–b). This first occurs in the inner core of southwest to south (Figures 2(h)–(i)).
the AR and successively proceeds to surrounding loops. In running ratio space-time plots at 193 (Figure 3,
The expansion experiences a gradual and then rapid ac- top), the diffuse pulse appears as a diagonal stripe of
celeration. Parabolicfits to its space-time position (Fig- emission increase (bright) followed by decrease (dark),
ure 3(e)) before and after 03:11 UT yield accelerations the boundary between which corresponds to the pulse
of 26±1 and 94±5ms−2 and final velocities of 52±3 peak. In Sector A4, for example, the peak coincides
and 234±13kms−1, close to the 286 kms−1 velocity of with the fast sharp front and propagates up to 800′′
the halo CME measured by SOHO LASCO. STEREO (580 Mm) from the flare kernel. At 171 (Figure 3, bot-
Ahead,67◦ eastofthe Earth,observesthiseventatnear tom), the pulse is less pronounced and, in contrast, ap-
pears as emission reduction (dark followed by bright;
quadrature and indicates that the lateral expansion of
Wills-Davey & Thompson 1999; Dai et al. 2010) rather
the CME is also primarily toward the south (Figure 1,
top; Movie 1). The similar velocities and common direc- than commonly observed enhancement (e.g., Long et al.
2008). Together, these observations suggest that the
tionsuggestthattheeruptingloopsarepartoftheCME
cooler 171 material is heated to 193 temperatures.
body.
First Global EUV “Wave” Observed by SDO AIA 3
Fig. 2.— SequentialAIAimagesat193and171˚A:(a)–(b)original(seeonlineMovies2a–b)and(c)–(i)runningdifference(Movies2c–d),
with two field of views (FOVs). Overlaid are two sets of spherical sectors, A0–A11 in (a) and B0–B8 in (h), used to obtain space-time
plots. Theinsertin(h)highlightsmultipleripples.
The side view fromSTEREO-Aindicates that the dif- proximatelyperpendicular to the localdisturbance front
fuse EUV enhancement is confined in the low corona (to which the A sectors are not).
and its leading front matches the lateral extent of ForselectedSectorsB0andB3,wecomposedbasedif-
the whitelight CME (Patsourakos & Vourlidas 2009; ferencespace-timeplots,normalizedbytheiraveragepre-
Kienreich et al. 2009). It is followed by an arc-shaped event brightness. We then obtained time profiles from
sharpincreaseandthendecreaseofemissionthatextends horizontalslicesateverydistanceofthe space-timeplot.
into the COR1 FOV (high corona). Its projection onto As shown in Figures 4(a) and (e) for 193, the profiles
the disk likely corresponds to the diffuse pulse’s peak representnet emission changes and clearly show a single
and/or one of the sharp fronts seen by AIA, according hump successively delayed with distance, corresponding
to its latitudinal position, sharpness, and large vertical tothepropagatingdiffusepulse. We definethepulse on-
extent (thus large LOS integration). set at the 20% level between the peak and the pre-event
minimum, from which the pulse amplitude and FWHM
4.2. Diffuse Pulse Shape Evolution arecalculated. Notethatthesetemporalprofilesatfixed
To best track the strongdisturbance in the southwest, locations areequivalent to spatialprofiles at giventimes
we used a finer set of 10◦ wide sectors (B0–B8, Fig- (i.e.,verticalslicesofspace-timeplots),becausethepulse
ure2(h))centeredatx=−275′′,y =306′′,whichareap- is fairlystable andpropagatesatconstantvelocities(see
4 Liu et al.
Fig.3.— Running ratio space-time plots obtained from A sectors shown inFigure 2(a) at 193 (top) and 171 (bottom), smoothed with
a60sboxcar. In(c)and(d), crosssignsconnected bydotted linesmarklinearfitstothepeakofthediffusepulseat193,withthefitted
velocitiesprintedonthetop. Theirendpointsarerepeatedinthecorresponding171panelsatthebottom. Otherdottedlinesareparabolic
(linearinFigure5)fitstovariousfeatures,withthegreateroftheinitialandfinalvelocitieslistedontheleftintheorderofthestarttime
ofthefit(markedbyanopencircle). Fitsin(e)arerepeated in(a). Movie3showsthisfigurewithandwithoutlabelsandfits.
Section 4.3). 4.3.1. Diffuse Pulse
In B0, where the slow sharp front is absent, the dif-
At 193 (Figures 4(b) & (f) and 5(a)–(d)), both the
fusepulseclearlydevelopsbeyonds =190′′(370′′south
0 onset and peak of the diffuse pulse travel at constant
of the flare kernel), within which it is not well-defined.
velocities in their mid stages, with the former being
Thepulseamplitudecanbefittedwithalinearfunction, slightly faster (e.g., 218±21 vs. 186±9kms−1 for Sec-
which decreases with distance by a fraction of 6 from
7 tor B0). This, despite the lack of a general trend for
a 41% enhancement at s to 6% at 675′′ (Figure 4(c)).
0 the overallpulse FWHM notedabove,indicatesthat the
(For comparison, the corresponding maximum emission
rising portion of the pulse broadens at increasing dis-
reduction at 171 is only ∼10%.) It decreases faster than
tances in all directions, consistent with that found by
the corresponding r−1 fit but slower than r−2, implying Veronig et al. (2010). In addition, the pulse onset veloc-
the pulse being neither a surface nor a spherical wave. ityweaklydepends ondirections,withanarrowrangeof
The pulse FWHM decreasesby only 1 from902to 581s 204–238kms−1.
3
(Figure4(d)),ratherthanincreasesasexpectedfromdis- At 171 (Figures 5(e)–(h)), with the signal being con-
persion of a wave. siderably weaker, the maximum emission reduction of
In B3, where the fast and slow sharp fronts overlap the diffuse pulse propagates at similar velocities as its
(see Movie 4b), the pulse shape is quite different. De- 193 counterpart of emission enhancement, e.g., 175±9
spite complications causedby stationarybrightenings or vs. 186±9kms−1 for Sector B0. There is a delay by
dimmings of small-scale loops, an increase followed by a ∼230 s in time or ∼42 Mm in space, which we specu-
decreaseofthepulseamplitudeisroughlyanti-correlated late could result from different wavelength response to
with variations of the FWHM in the distance range the convolutionoftemperature and density effects, both
marked by the dashed lines (Figures 4(g)–(h)). This in- being likely present here.
dicatessteepeningandthenbroadeningofthepulse. The
narrowestpulse (near s=350′′) shows a sharp dropand 4.3.2. Sharp Fronts
deep, long-lasting dimming on its trailing edge, which
At193,thesharpfrontsarerecognizedasnarrowridges
are different from leading edge steepening in shock for-
inrunningratiospace-timeplots(Figures5(b)–(d)),and
mations.
someofthemexperience acceleration(e.g.,Figure5(d)).
Theirfinalvelocitiesindifferentdirectionsvarybyafac-
tor of two (136–238kms−1), in contrast to the narrow
4.3. Kinematics of Disturbance Propagation
range of the diffuse pulse’s onset velocity.
We derive kinematics of the disturbance propagation At 171 (Figure 5, bottom), most of the sharp fronts
from base and running ratio space-time plots for B sec- areemission enhancements,opposite to the diffuse pulse
tors as shown in Figure 5. reduction, but the same as their 193 counterparts at
First Global EUV “Wave” Observed by SDO AIA 5
Fig. 4.— Top: (a)Horizontalcutsofthenormalized193basedifferencespace-timeplotofSector B0showninFigure2(h). Eachcurve
represents the percentage emissionvariationatafixed location markedby thedotted line(alsorepresenting thepre-event level). (b)–(d)
Distancevs.onsetandpeaktimes,amplitude,andFWHMofthepulseshownin(a). Theredsolidlinesarelinearfits,andthereddashed
and dotted linesin(c) arer−1 and r−2 fits starting withthe same value at s=190′′. Vertical dotted lines markthe mean values inthe
distancerangeofinterest. Theorange“×”sin(c)–(d) indicatestationary brighteningsofsmallscaleloops. Bottom: sameasTopbutfor
Sector B3.
similar velocities (e.g., 128 ± 6 vs. 136 ± 7kms−1 for 4.4. Component Interaction: Crossing and Parallel
B4). This suggests a density increase at both channels’ Ripples
temperatures. The slow sharp front here is particularly
Near03:24UT,thefastsharpfront,travelingtwotimes
pronounced and dominant over the fast front (see Fig-
faster,overtakesthe slowsharpfrontinitiallyaheadofit
ures 5(f)–(h), inserts). Its velocity is only ∼50% that of
at a shallow angle (see Figure 2(h), Movies 4b–c). They
the diffuse pulse’s peak (e.g., 79±4 vs. 154±8kms−1 appear as crossingsharp ridges in space-time plots (Fig-
for B3) and decreases toward the west, giving rise to its
ure 5, inserts) and both propagate independently after-
southward elongated oval shape (Figure 2(f)). wards.
Incidentally, there are numerous slow (∼20kms−1), Two additional weaker, sharp fronts (“ripples”) are
outwardmovingfeatures(B6,Figure5), whicharelikely generated ahead of the original fast front (Figures 2(h)
theflanksoftheexpandingloopsmentionedearlier. The and 5(c)). Early in their lifetime the additional fronts
onsets of these features and the slow sharp front occur accelerate, asymptotically approaching the constant ve-
withinminutesofthearrivalofthediffusepulse’sleading locityoftheoriginalfront,whiletheirspatialseparations
edgeatincreasingdistances,suggestingtheformerbeing decrease with time. Afterwards, they propagate at the
triggered by the latter. sameconstantvelocityof∼130kms−1forupto∼10min-
utes,maintainingaconstantseparationof∼160sintime
or ∼20 Mm in space. This the first time such fine rip-
6 Liu et al.
Fig.5.— SameasFigure3butforbase(column1)andrunning(columns2–4)ratiosfromBsectorsshowninFigure2(h). Plus(cross)
signsconnected bydottedlinesarelinearfitstotheonset(peak)ofthediffusepulse. Theirendpointsfor193arerepeatedfor171atthe
bottom. Thecolorbarsontherightareforcolumn1only,whilecolumns 2–4sharethecolorscales withFigure3. Thegrayscaleinserts
provide enlarged views showing parallel and crossing ripples. (See online Movies 4a–c for this figure and running difference images from
whichthespace-timeplotsareobtained.)
ples are observedandthey aredifferent fromthose more direction of the CME expansion, which likely pro-
widelyseparated(byfractionsof R⊙)HeI“wave”pulses duces the strongest compression.
(Gilbert & Holzer 2004, their Fig. 2).
2. Somesharpfrontsexhibitaccelerationsandtheirfi-
5. DISCUSSION nalvelocities(e.g.,238±12kms−1,Figure5(d))are
The first SDO AIA observations of a global EUV dis- closetothoseoftheeruptingloopsandCME(Sec-
turbancepresentedherehaverevealedapictureofmulti- tion 3). Such accelerations (Zhukov et al. 2009),
ple temperatures and spatio-temporal scales in unprece- not expected for freely propagating MHD waves,
dented detail. likely reflect the acceleration of the CME in the
low corona (Zhang et al. 2001).
5.1. Separation of Sharp Fronts from Diffuse Pulse
3. The emission profile steepens on the trailing edge
AIA’s high cadences and sensitivities allowed us to
ofthe sharpfront(Figure 4(e)), suggestiveofcom-
discover and separate a new feature, the sharp fronts,
pression by the CME from behind. It is followed
fromthe concurrent,commonly seen diffuse pulse. With
bydeepdimming,anotherindicatoroftheexpand-
overtaking fast and slow components and no Moreton
ingCMEbody. Suchcompressionisalsoconsistent
waveassociation,thesharpfrontsseemtodifferfrom“S-
withthedensityenhancementsuggestedbybright-
waves”, a minority (∼7% of 173 events; Biesecker et al.
ening of the sharp fronts at both 193 and 171.
2002) of “EIT waves”, that are sharp, fast, and often
cospatial with Moreton waves (Thompson et al. 2000). 4. The leading edge of the diffuse pulse travels
at constant velocities with very little direction-
5.2. Hybrid Wave and Non-wave Hypothesis
dependence. This is expected for a fast mode
We propose a hybrid hypothesis combining both wave MHD wave. Its velocity of 204–238kms−1
and non-waveaspects to best, although not entirely sat- (cf., 500kms−1, Harra & Sterling 2003) is roughly
isfactorily, explain the observations. Namely, the sharp within the estimated range of fast mode speeds
bright fronts are caused by CME compression traveling on the quiet Sun (215–1500kms−1, Wills-Davey
behindtheweaker,moreuniformdiffusefrontwhichisan 2006). Inthe absenceofa shock,this speedshould
MHD wave generated by the CME. This is conceptually begreaterthanthatofthewavegenerator(i.e.,the
similar to the wave/non-wave bimodality suggested by CME preceded by the sharp fronts), as observed
Zhukov & Auch`ere(2004)andsimulatedbyCohen et al. here.
(2009). Our specific interpretations are as follows.
5. The linear drop of the pulse amplitude and the
1. The bright sharp fronts are located to the south- narrowing of the pulse width with distance (Sec-
west of the erupting AR, consistent with the main tion 4.2; cf. Veronig et al. 2010) are inconsistent
First Global EUV “Wave” Observed by SDO AIA 7
withanyfreelypropagatingMHDwaves,butmight in similar directions. The velocities of the slow sharp
be explained by the superposition of the CME- frontaretoosmallforfastmodewaves,whileslowmode
driven wave and non-wave components. waves are not favorable candidates either because they
cannot propagate perpendicular to magnetic fields.
5.3. Open Questions Another problem with the fast mode interpretation is
the smaller than expected “wave” velocity which was
The main challenge to the above hybrid interpreta-
ascribed to undersample by low instrument cadences
tion is the interaction of the fast and slow sharp fronts
(Long et al.2008). AIA’shighcadencesherehavenotre-
(Section 4.4), which shows many characteristics of real
vealedmuchanticipatedhighervelocities,anditremains
waves. The overtaking fronts can be explained by two
tobe seenifthisappliestomoreenergeticevents. Inad-
wavepulseswithdifferentvelocities. Whenafasterpulse
dition,thediffusepulse’svelocityremainsconstant(con-
approaches and overtakes a slower pulse, their spatial
sistent with earlier findings; e.g., White & Thompson
superimposition naturally leads to the observed anti-
2005; Ma et al. 2009), rather than decelerates like a
correlation of the pulse amplitude and FWHM (Fig-
shock (not necessarily true for “EIT waves”) degener-
ure 4(g)–(h)). They then propagate independently af-
ating to a fast mode wave (Warmuth 2007).
terwards, as observed here. However, if a sharp front
The nature of global EUV disturbances thus remains
represents a layer/loop of the erupting CME structure,
elusive. However, this is just the beginning, and we an-
a faster layer would sweep and compress a slower layer
ticipate that continuing analysis of new data from AIA,
ahead of it to form a single layer of stronger compres-
STEREO,andotherinstrumentswillprovidemorecriti-
sion. This would not lead to overtaking,unless they are
cal information and help eventually end the decade-long
at different altitudes and LOS projection gives an im-
debate.
pression of overtaking, for which we found no evidence
from STEREO observations.
Meanwhile, this also challenges pure wave models. If
thefastandslowsharpfrontsareofthesameMHDwave This work was supported by AIA contract
mode, they would propagate at the same characteristic NNG04EA00C. We thank the referee for construc-
speed dictated by the medium, contradicted by the fact tive comments and Richard Shine, Mei Zhang, and the
thattheirobservedvelocitiesdifferbyafactoroftwoeven SDO team for their help.
REFERENCES
Attrill,G.D.R.2010,ApJ,718,494 Thompson,B.J.,Reynolds,B.,Aurass,H.,Gopalswamy,N.,
Attrill,G.D.R.,Harra,L.K.,vanDriel-Gesztelyi,L.,& Gurman,J.B.,Hudson,H.S.,Martin,S.F.,&St.Cyr,O.C.
D´emoulin,P.2007,ApJ,656,L101 2000,Sol.Phys.,193,161
Biesecker,D.A.,Myers,D.C.,Thompson,B.J.,Hammer,D.M., Veronig,A.M.,Muhr,N.,Kienreich,I.W.,Temmer,M.,&
&Vourlidas,A.2002, ApJ,569,1009 Vrˇsnak,B.2010, ApJ,716,L57
Chen,P.F.,Fang,C.,&Shibata,K.2005,ApJ,622,1202 Veronig,A.M.,Temmer,M.,&Vrˇsnak,B.2008, ApJ,681,L113
Chen,P.F.,Wu,S.T.,Shibata,K.,&Fang,C.2002,ApJ,572, Vrˇsnak,B.&Cliver,E.W.2008, Sol.Phys.,253,215
L99 Wang,H.,Shen,C.,&Lin,J.2009,ApJ,700,1716
Cohen,O.,Attrill,G.D.R.,Manchester,W.B.,&Wills-Davey, Wang,Y.2000,ApJ,543,L89
M.J.2009,ApJ,705,587 Warmuth,A.2007,inLectureNotes inPhysics,BerlinSpringer
Dai,Y.,Auch`ere,F.,Vial,J.,Tang,Y.H.,&Zong, W.G.2010, Verlag,ed.K.-L.Klein&A.L.MacKinnon,Vol.725,107–138
ApJ,708,913 Warmuth,A.,Vrˇsnak, B.,Aurass,H.,&Hanslmeier,A.2001,
Delann´ee,C.2000, ApJ,545,512 ApJ,560,L105
Gallagher,P.T.&Long,D.M.2010,submittedtoSpaceScience White,S.M.&Thompson,B.J.2005,ApJ,620,L63
Reviews Wills-Davey,M.J.2006,ApJ,645,757
Gilbert,H.R.&Holzer,T.E.2004,ApJ,610,572 Wills-Davey,M.J.&Attrill,G.D.R.2009,SpaceScience
Harra,L.K.&Sterling,A.C.2003, ApJ,587,429 Reviews,149,325
Kienreich,I.W.,Temmer,M.,&Veronig,A.M.2009,ApJ,703, Wills-Davey,M.J.,DeForest,C.E.,&Stenflo, J.O.2007,ApJ,
L118 664,556
Long,D.M.,Gallagher,P.T.,McAteer,R.T.J.,&Bloomfield, Wills-Davey,M.J.&Thompson,B.J.1999, Sol.Phys.,190,467
D.S.2008,ApJ,680,L81 Wu,S.T.,Zheng,H.,Wang,S.,Thompson,B.J.,Plunkett,S.P.,
Ma,S.,Wills-Davey,M.J.,Lin,J.,Chen,P.F.,Attrill,G.D.R., Zhao,X.P.,&Dryer,M.2001,J.Geophys. Res.,106,25089
Chen,H.,Zhao,S.,Li,Q.,&Golub,L.2009,ApJ,707,503 Zhang,J.,Dere,K.P.,Howard,R.A.,Kundu,M.R.,&White,
Moreton,G.E.1960, AJ,65,494 S.M.2001,ApJ,559,452
Moses,D.,etal.1997,Sol.Phys.,175,571 Zhukov,A.N.&Auch`ere,F.2004,A&A,427,705
Ofman,L.&Thompson,B.J.2002,ApJ,574,440 Zhukov,A.N.,Rodriguez,L.,&dePatoul,J.2009,Sol.Phys.,
Patsourakos,S.&Vourlidas,A.2009,ApJ,700,L182 259,73
Podladchikova, O.&Berghmans,D.2005,Sol.Phys.,228,265
Schmidt,J.M.&Ofman,L.2010, ApJ,713,1008
Thompson,B.J.&Myers,D.C.2009, ApJS,183,225
Thompson,B.J.,Plunkett,S.P.,Gurman,J.B.,Newmark,J.S.,
St.Cyr,O.C.,&Michels,D.J.1998, Geophys.Res.Lett.,25,
2465